Radiotherapy in the treatment of locoregional ... - BIR Publications

The British Journal of Radiology, 75 (2002), 663–669

E

2002 The British Institute of Radiology

Radiotherapy in the treatment of locoregional relapses of breast cancer 1

¨ NEMANN, MD, 1B MATTHEES, 2C E RU ¨ BE, MD, 1G REINARTZ, MD, A SCHUCK, MD, 1S KO 1 1 S HESSELMANN, MD, O MICKE, MD, U SCHA¨FER, MD and 1N WILLICH, MD

1 1 2

Radiotherapy Department, University of Muenster, Albert-Schweitzer-Str. 33, D-48129 Muenster and Radiotherapy Department, University of Homburg/Saar, 66421 Homburg/Saar, Germany

Abstract. Locoregional recurrences of breast cancer are associated with considerable morbidity and frequently present with concurrent metastatic disease. Yet patients without systemic spread can be treated with curative intent. In a retrospective analysis, the results of treatment of these patients have been evaluated at our institution. Between 1987 and 1996, 113 patients with locoregional breast cancer relapse, without systemic manifestation, received irradiation after local tumour excision. 13 patients (11.5%) had already received radiotherapy as part of their primary treatment. In these cases, only the area involved was treated. In all other patients, the chest wall and the ipsilateral lymph nodes were irradiated. Median dose was 50 Gy (range 20–65 Gy). Median follow-up was 4.4 years. 76 patients (67.3%) presented with chest wall recurrence only, 25 patients (22.1%) with nodal relapse only and 12 patients (10.6%) with combined relapses. 93% of patients had local control of disease after treatment. Local control rate after 5 years was 59%. 63 patients (55.8%) died within the follow-up interval, 45 patients (39.8%) owing to metastases, 4 patients (3.5%) owing to local failure and 8 patients (7%) owing to causes unrelated to tumour. Overall survival after 5 years was 43%. In multivariate analysis, positive hormone receptor status, small tumours on relapse and chest wall relapses alone were associated with improved survival. Radical local therapy is necessary in order to achieve and maintain local control and to prevent secondary dissemination in patients with only local recurrence of breast cancer.

With either breast conserving therapy, including post-operative radiotherapy or modified radical mastectomy, the local relapse rates after 5 years are approximately 5–10%, although local recurrence can be as high as 30% in high-risk patients [1–3]. Although purely locoregional failures can be treated with curative intent, morbidity of local relapse is considerable and often accompanied or followed by systemic failure [4–8]. The goal of local treatment is to avoid further local relapse and to prevent secondary dissemination in patients who have not developed systemic spread [9–12]. In this retrospective analysis, the outcome of patients with locoregional relapse of breast cancer has been evaluated.

Materials and methods Between 1987 and 1996, 173 patients with locoregional relapse of breast cancer were treated at our institution. Local relapse was defined as relapse on the chest wall, and regional relapse included the axilla, the supra- or infraclavicular fossa and the parasternal region. 113 patients Received 12 April 2001 and in final form 6 February 2002, accepted 8 February 2002. The British Journal of Radiology, August 2002

presented with locoregional relapse, including failures in the supraclavicular fossa, without evidence of further dissemination. Analysis focuses on this subgroup of patients with a favourable prognosis. Treatment was dependent on the therapy of the primary tumour and on the site of recurrence. Initial surgical procedures are shown in Table 1. 13 patients (11.5%), including patients with quadrantectomy and tylectomy, had received post-operative radiotherapy with a cobalt-60 unit as part of primary treatment. The remainder had not received post-operative radiotherapy. 65 patients (57.5%) had received adjuvant hormonal or cytotoxic therapy. All patients with chest wall recurrence alone underwent excision of the recurrence, 20 of 25 patients with regional relapse alone underwent resection of the involved nodes, 3 patients underwent biopsy only and 2 patients were irradiated without histological confirmation. There was only one patient who developed parasternal nodal relapse. 6 of 12 patients with combined chest wall and nodal relapses underwent complete resection of the locoregional disease, 6 patients underwent biopsy only. Of the 22 patients with initial breast conserving 663

A Schuck, S Ko¨nemann, B Matthees et al Table 1. Surgical procedure for primary tumour Initial surgery

No.

Radical ME Modified radical ME Subcutaneous ME Quadrantectomy Tylectomy

1 88 14 5 3

ME, mastectomy.

therapy, 9 were treated by mastectomy, 1 by subcutaneous mastectomy and 1 by quadrantectomy. 10 patients underwent local excision and 1 patient a biopsy. All patients who had received no previous irradiation received radiotherapy to the chest wall and the axillary, supra- and infraclavicular lymph nodes. Patients who had initially been irradiated received local irradiation of the areas of relapse only, with a margin of 2 cm around the lesion. Median dose was 50 Gy (range 20–65 Gy). Follow-up data were obtained from the patients’ follow-up examinations at our institution or by the local gynaecologist. Minimum follow-up was 2 years and median follow-up 4.4 years (range 2.0–11.4 years). Life tables were calculated using the Kaplan– Meier method and univariate analyses were calculated using the log-rank test. Multivariate analysis was performed using Cox regression [13]. For locoregional control, locoregional relapses were the only events considered; death or the end of follow-up were censored. For overall survival (OAS), death for any reason was considered as an event whilst the end of follow-up was censored.

Results At diagnosis of the primary tumour, 21 patients (18.6%) had stage I disease, 49 patients (43.4%)

had stage II disease and 21 patients (18.6%) stage III disease. In 22 patients (19.5%), the initial T or N stage was not documented. 44.2% of patients had histologically positive and 36.3% histologically negative ipsilateral lymph nodes. For the other patients, pathological lymph node status was not available. All patients were free from evidence of distant metastases. Of the 86 patients (76.1%) with known hormone receptor status at diagnosis of the primary tumour, 42 had a positive oestrogen and 35 a positive progesterone receptor status. The most frequent location of locoregional recurrence was the chest wall alone, affecting 76 patients (67.3%). In 70 of these 76 patients, the scar of the primary operation was involved. 25 patients (22.1%) had a regional relapse only, and 12 patients (10.6%) had a combined nodal and chest wall relapse. The size of recurrence was ,1 cm in 43 patients (38.1%), 1–3 cm in 53 patients (46.9%) and .3 cm in 17 patients (15.0%). 80% of local relapses occured within the first 5 years after diagnosis of the primary tumour, the latest local relapse occuring after an interval of 20.7 years. At first follow-up examination, 2 months after completion of radiotherapy, 93% of recurrences were locally controlled, local control rate after 5 years being 59%. After 7 years, no further local relapses occured. Local control rate is shown in Figure 1. Factors influencing local control in univariate analysis are shown in Table 2. 63 patients (55.8%) died within the follow-up interval, 45 patients (71.4%) from metastases, 4 patients (6.3%) of local failure and 8 patients (12.7%) for reasons not related to cancer. Cause of death in 6 patients (9.6%) was not known. OAS after 5 years for the entire group was 43% (Figure 2). In uni- and multivariate analyses,

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

664

2

4

6

8

10

12

Figure 1. Locoregional control after therapy of locoregional recurrence (n5113).

The British Journal of Radiology, August 2002

Locoregional relapses of breast cancer Table 2. Risk factors for local failure in univariate analysis Risk factor

Significance

Initial advanced T stage Initial negative hormone receptor status status Relapse free interval ,2 years Nodal or combined nodal and chest wall relapse wall replase Initial node positive disease Previous irradiation Size of local relapse Menopausal status

p50.002 p50.012 p50.002 p50.001 p50.078 p50.54 p50.14 p50.51

Table 3. Risk factors for survival Risk factors

Univariate analysis

Multivariate analysis

Initial negative hormone receptor status Relapse free interval ,2 years Local relapse .1 cm Nodal or combined nodal and chest wall relapse Previous irradiation Initial advanced T stage Initial node positive disease Menopausal status

p50.0001

p50.0011

p50.0006

n.s.

p50.0079 p50.0002

p50.0043 p50.0068

p50.54 p50.21 p50.06

— — —

p50.46

—

n.s., not significant.

survival was influenced by features of both primary tumour and local relapse (Table 3). OAS of patients with different sites of relapse is illustrated in Figure 3. With doses up to 40 Gy, local control was

not achieved in 5 (62.5%) of 8 patients. The corresponding failure rate was 23 (34.3%) of 67 patients for 41–50 Gy, 13 (40.6%) of 32 patients for 51–60 Gy and 3 (50%) of 6 patients above

1.0

0.8

0.6

0.4

0.2

0 0

4

2

8

6

10

12

Figure 2. Overall survival locoregional relapse (n5113).

after

1.0

0.8

0.6

0.4

0.2

0 0

2

4

6

8

10

12

Figure 3. Overall survival depending on relapse site (p50.0002). The British Journal of Radiology, August 2002

665

A Schuck, S Ko¨nemann, B Matthees et al

60 Gy. Four patients received much less than 40 Gy. One patient with an axillary recurrence had already received 60 Gy to the axilla and therefore received only a further 20 Gy. Three patients who received 25 Gy, and should have received a second course of 25 Gy, deteriorated and so radiotherapy was stopped. Erythema was recorded in 34 patients, moist desquamation in 10 patients, oedema of the arm with subjective impairment in 5 patients and reduced mobility of the arm in 3 patients.

Discussion Patients with locoregional recurrence of breast cancer without concurrent systemic disease can be treated with curative intent [11, 12, 14]. Although a considerable number of patients will subsequently develop systemic failure, radical locoregional therapy for locoregional relapses is given to avoid further local recurrence and secondary spread. In our analysis, local control after 5 years was 57% (Figure 1) and OAS was 43% (Figure 2). The treatment of local relapse after mastectomy should include complete excision of the tumour if possible [15] and post-operative radiotherapy. Local relapse after initial breast conserving therapy including adjuvant irradiation should be treated with ablation; additional post-operative irradiation is reserved for patients with incomplete resection. Retrospective analyses, reporting treatment results with and without irradiation for local relapse after initial mastectomy, show an increased rate of subsequent local failures in patients that do not receive irradiation after excision of local relapse [5, 10, 16]. The impact of irradiation on survival is not clear due to small patient numbers and inhomogeneous patient populations. There are several reports of chest wall resections alone in patients with advanced locally recurrent breast cancer [17–20]. Patients who are eligible for this extensive procedure need to have good performance status without concurrent metastatic disease, localized infiltrating recurrence of the chest wall and a long disease-free interval indicating a favourable prognosis. The addition of radiation after extensive surgery may be appropriate depending on individual tumour and treatment characteristics. In patients with local relapse who have received no previous radiation to the chest wall, the entire ipsilateral chest wall should be irradiated. Several authors report improved local control, and even survival, after irradiation of the entire chest wall compared with small fields [7, 10, 11, 21]. In Halverson et al’s report, freedom from further chest wall recurrence after 10 years was 63% for 666

patients who received irradiation to the whole chest wall but only 18% for those who did not (p50.0001) [11]. There are data from several reports that patients with chest wall relapse alone who also receive adjuvant irradiation to the ipsilateral supraclavicular fossa have better subsequent locoregional control [10–12, 22, 23]. There is no evidence that these patients benefit from adjuvant irradiation of the ipsilateral axilla and little evidence that they would benefit from irradiation of the internal mammary lymph nodes. Furthermore, patients with lymph node relapse have better local control after adjuvant irradiation of the ipsilateral chest wall [10–12]. Some investigators report improvement in survival in patients who received adjuvant irradiation of uninvolved sites, but this difference could reflect a selection bias in the groups compared [11, 22]. Although a dose–response relationship above 40 Gy could not be established from our data, as well as from some reports in the literature [7, 24], an adequate dose of radiation should be given in an adjuvant setting as well as for macroscopic residual disease. Bedwinek et al [10] observed that excised local relapses could be controlled with doses of 50 Gy or above. Small, unresected local relapses could be controlled with doses .55 Gy. Halverson et al reported that subclinical disease could be controlled with 50–60 Gy [11]. Small incompletely resected tumours of less than 3 cm diameter should be irradiated with 60 Gy or more. Only 50% of large tumours could be controlled with doses up to 70 Gy. Patients who have received previous radiotherapy to the chest wall do not usually receive another full course of treatment to the entire chest wall because of associated toxicity. Stadler et al [7] and Haylock et al [4] did not find that previous irradiation of the primary tumour influenced local control. Toonkel et al [22] observed an increased rate of locoregional failures due to small radiation fields. No author reported an influence of previous irradiation on survival [4, 14, 22]. In our own series, the 13 patients (11.5%) with previous irradiation of the site of relapse did not show decreased local control or survival. Although an increase in locoregional failure rate would not be unexpected because of the reduced field sizes used after previous radiation, the magnitude of the effect seems to be limited. In local relapse after breast conserving therapy, mastectomy is recommended if possible. A number of retrospective analyses of mastectomy after local relapse of conservatively managed breast cancer reported better results than the series of patients with local relapse after initial mastectomy [25–28]. This can be partly explained by the The British Journal of Radiology, August 2002

Locoregional relapses of breast cancer

selection of a subgroup of patients who have localized operable relapse. In a recent publication, Tienhoven et al reported the outcome of patients with local relapse alone in the EORTC trial 10801 and the Danish breast cancer trial DBCG-82TM [8]. In both trials patients were randomized to receive breast conserving therapy, including post-operative irradiation or mastectomy. No difference in local control and survival could be detected between the treatment groups after local relapse. The 5 year OAS was approximately 60% in each group. In 27% of the patients initially treated conservatively, salvage mastectomy was not possible. Haffty et al [29] observed a 59% 5 year OAS following local relapse after initial breast conserving therapy as well. These reports demonstrate that there is probably only a small, or no, prognostic difference in local relapse after breast conserving therapy or mastectomy. A number of reports have dealt with risk factors for local control and survival after locoregional recurrence of breast cancer. Risk factors for locoregional control in univariate or multivariate analyses are initial positive nodal status, initial advanced T stage, negative hormone receptor status, relapse-free interval ,2 years, incomplete or no excision of the relapsed tumour, large or multiple lesions, nodal relapses and omission of adjuvant irradiation of the supraclavicular fossa and the chest wall [4, 5, 7, 8, 10–12, 22, 30]. In our own series, initial advanced T stage, initial negative oestrogen and progesterone receptor status, relapse-free interval ,2 years and nodal or combined nodal and chest wall relapse were associated with poor local control (Table 2). Risk factors for survival in univariate or multivariate analyses are initial positive nodal status, initial large T stage, initial relapse-free interval ,2 years, nodal or combined nodal and chest wall recurrences, large or multiple lesions, vascular invasion, incomplete or no excision of the recurrent tumour and local failure after therapy for relapse [4–8, 10–12, 21, 22, 29, 30– 33]. In our own univariate analysis, a negative initial hormone receptor status, a relapse-free interval ,2 years, nodal or combined nodal and chest wall recurrence and local relapse .1 cm were associated with poor survival. There was a trend for decreased survival in patients who were initially node positive compared with those who were initially node negative (p50.06). In multivariate analysis, receptor status, relapse tumour size and tumour location remained significant (Table 2). There have been no published randomized series comparing local treatment and local treatment plus chemotherapy in locoregional relapse. There have been a number of reports that The British Journal of Radiology, August 2002

retrospectively evaluated the impact of systemic therapy, including one matched pair analysis. Data were not conclusive because of biases in the different groups and the application of different regimens and numbers of cycles of chemotherapy [5, 14, 34]. Haylock et al [4] performed a prospective, but not randomized, study comparing local therapy alone (excision if feasible and radical locoregional radiotherapy) with local therapy and chemotherapy (eight cycles of doxorubicin and cyclophosphamide). Hormone therapy was given to all patients with positive oestrogen receptor status. The 61 patients treated from 1979 to 1983 received chemotherapy and showed a non-significant trend towards improved breast cancer specific survival, distant recurrencefree survival and OAS compared with the 59 patients treated from 1983 to 1989 who did not receive chemotherapy. There was no difference in locoregional control. A prospective randomized trial of hormone therapy after locoregional relapses in breast cancer was performed in Switzerland. Borner et al randomized low risk patients (R0/1 resection, oestrogen receptor positive in the biopsy of the recurrence or, if unknown, relapse-free interval .1 year, less than 4 nodules, largest nodule ,3 cm) to radiation alone or radiation plus 20 mg tamoxifen per day until progression [31]. There was a significantly improved relapse-free survival in the combined treatment group owing to significantly improved locoregional control. There was no improvement in the control of systemic disease and OAS was identical in the two groups. Since relapse-free survival and local control are relevant for patients despite unaltered OAS, tamoxifen is recommended in patients with positive oestrogen receptor status if not given initially. These reports underline the need for prospective randomized trials to define the role of chemotherapy after local relapse.

Conclusion Intensive local therapy is necessary for isolated locoregional relapse of breast cancer in order to obtain and maintain local control and prevent secondary dissemination. After excision and postoperative irradiation, 5-year local control rate in our series was 59% and OAS was 43%. Initial advanced T stage, initial negative hormone receptor status, relapse-free interval ,2 years and nodal or combined nodal and chest wall relapses were risk factors for poor local control. Initial negative hormone receptor status, large local relapses and nodal or combined nodal and chest wall relapses were risk factors for poor survival in multivariate analysis. 667

A Schuck, S Ko¨nemann, B Matthees et al

References 1. Overgaard M, Jensen MB, Overgaard J, Hansen P, Rose C, Andersson M, et al. Postoperative radiotherapy in high risk postmenopausal breast cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82C randomised trial. Lancet 1999;353:1641–8. 2. Recht A, Gray R, Davidson N, Fowble B, Solin L, Cummings F, et al. Locoregional failure 10 years after mastectomy and adjuvant chemotherapy with or without irradiation: experience of the Eastern Cooperative Oncology Group. J Clin Oncol 1999; 17:1689–1700. 3. Veronesi U, Luini A, Galimberti V. Conservation approaches for the management of stage I/II carcinoma of the breast. Milan cancer institute trials. World J Surg 1994;18:70–5. 4. Haylock B, Coppin C, Jackson J, Basco VE, Wilson KS. Locoregional first recurrence after mastectomy: prospective cohort studies with and without immediate chemotherapy. Int J Radiat Oncol Phys 2000;46:355–62. 5. Janjan N, McNeese M, Buzdar A, Montague E, Oswald M. Management of locoregional recurrent breast cancer. Cancer 1986;58:1552–6. 6. Kamby C, Sengelov L. Pattern of dissemination and survival following isolated locoregional recurrence of breast cancer. Breast Cancer Res Treat 1997; 45:181–92. 7. Stadler B, Kogelnik D. Local control and outcome of patients treated for isolated chest wall recurrence of breast cancer. Radiother Oncol 1987; 8:105–11. 8. Van Tienhoven G, Voogd AC, Peterse JL, Nielsen M, Andersen KW, Mignolet F, et al. Prognosis after treatment for loco-regional recurrence after mastectomy or breast conserving therapy in two randomised trials (EORTC 10801 and DBCG82TM). Eur J Cancer 1999;35:32–8. 9. Aberizk W, Silver B, Henderson C, Kady B, Harris JR. The use of radiotherapy for treatment of isolated locoregional recurrence of breast carcinoma after mastectomy. Cancer 1986;58:1214–8. 10. Bedwinek J, Fineberg B, Lee J. Analysis of failure following local treatment of isolated local-regional recurrence of breast cancer. Int J Radiat Oncol Biol Phys 1981;7:581–5. 11. Halverson K, Perez C, Kuske R, Garcia D, Simpson J, Fineberg B. Isolated local-regional recurrence of breast cancer following mastectomy: radiotherapeutic management. Int J Radiat Oncol Biol Phys 1990;19:851–8. 12. Schwaibold F, Fowble B, Solin L, Schultz D, Goodman R. The results of radiation therapy for isolated local regional recurrence after mastectomy. Int J Radiat Oncol Biol Phys 1991;21: 299–310. 13. Cox DR. Regression models and lifetables. J Roy Stat Soc 1972;B34:187–220. 14. Deutsch M, Parsons J, Mittal B. Radiation therapy for local-regional recurrent breast carcinoma. Int J Radiat Oncol Biol Phys 1986;12:2061–5. 15. Winzer KJ, Guski H, Aldinger HD, Mu¨ller JM. Das lokoregiona¨re Rezidiv beim MammakarzinomDiagnostik und Therapie. Akt Chir 1996;31:113–8. 16. Eek R, Falkson C. Extended survival in 80 patients without operable locoregionally recurrent breast cancer treated with chemotherapy. Am J Clin Oncol 1998;21:501–4. 668

17. Faneyte I, Rutgers E, Zoetmulder F. Chest wall resection in the treatment of locally recurrent breast carcinoma. Cancer 1997;80:886–91. 18. Hathaway C, Rand R, Moe R, Machioro T. Salvage surgery for locally advanced and locally recurrent breast cancer. Arch Surg 1994; 129:582–7. 19. Kluiber R, Bines S, Bradley C, Faber LP, Witt TR. Major chest wall resection for recurrent breast carcinoma. Am Surg 1991;57:523–9. 20. Miyauchi K, Koyama H, Noguchi S, Ianji H, Yamamoto H, Kodama K, et al. Surgical treatment for chest wall recurrences of breast cancer. Eur J Cancer 1992;28:1059–62. 21. Willner J, Kiricuta C, Ko¨lbl O. Locoregional recurrence of breast cancer following mastectomy: always a fatal event? Results of univariate and multivariate analysis. Int J Radiat Oncol Biol Phys 1997;37:853–63. 22. Toonkel L, Fix I, Jacobson L, Wallach C. The significance of local recurrence of carcinoma of the breast. Int J Radiat Oncol Biol Phys 1983;9: 33–9. 23. Chen K, Montague E, Oswald M. Results of irradiation in the treatment of locoregional breast cancer recurrence. Cancer 1985;56:1269–73. 24. Mendenhall N, Devine J, Mendenhall W, Bland K, Million R, Copeland E. Isolated local-regional recurrence following mastectomy for adenocarcinoma of the breast treated with radiation therapy alone or combined with surgery and/or chemotherapy. Radiother Oncol 1988;12:177–85. 25. Abner A, Recht A, Eberlein T, Come S, Shulman L, Hayes D, et al. Prognosis following salvage mastectomy for recurrence in the breast after conservative surgery and radiation therapy for early stage breast cancer. J Clin Oncol 1993;11: 44–8. 26. Fowble B, Solin L, Schultz D, Rubenstein J, Goodman R. Breast recurrence following conservative surgery and radiation: patterns of failure, prognosis, and pathologic findings from mastectomy specimens with implications for treatment. Int J Radiat Oncol Biol Phys 1990;19: 833–42. 27. Kempermann H, Borger J, Hart A, Peterse H, Bartelink H, Van Dongen J. Prognostic factors for survival after breast conserving therapy for stage I and II breast cancer. The role of local recurrence. Eur J Cancer 1995;31:690–8. 28. Kurtz J, Amalric R, Brandone H, Ayme Y, Jacquemier J, Pietra JC. Local recurrence after breast-conserving surgery and radiotherapy. Cancer 1989;63:1912–7. 29. Haffty B, Fischer D, Beinfield M, McKhann C. Prognosis following local recurrence in the conservatively treated breast cancer patient. Int J Radiat Biol Phys 1991;21:293–8. 30. Magno L, Bignardi M, Micheletti E, Bardelli D, Plebani F. Analysis of prognostic factors in patients with isolated chest wall recurrence of breast cancer. Cancer 1987;60:240–4. 31. Borner M, Bacchi M, Goldhirsch A, Harder F, Castiglione M, Jungi W, et al. First isolated locoregional recurrence following mastectomy for breast cancer: results of a phase III multicenter study comparing systemic therapy with observation after excision and radiation. J Clin Oncol 1994; 12:2071–7. The British Journal of Radiology, August 2002

Locoregional relapses of breast cancer 32. Brunner K, Harder F, Greiner R, Frost M, Cavalli F, Senn H, et al. Das loko-regionale rezidiv nach operiertem mammakarzinom: prognostische faktoren und therapeutische konsequenzen. Schweiz Med Wochenschr 1988;118:1976–81. (In German.) 33. Gilliland M, Barton R, Copeland E. The implications of local recurrence of breast cancer as the first site of therapeutic failure. Ann Surg 1983; 197:284–7.

The British Journal of Radiology, August 2002

34. Halverson K, Perez C, Kuske R, Garcia D, Simpson J, Fineberg B. Locoregional recurrence of breast cancer: a retrospective comparison of irradiation alone versus irradiation and systemic therapy. Am J Clin Oncol 15 1992:93–101.

669